Comparative agreement of clinical diagnoses and neonatal autopsy findings in a tertiary care center

Aim: To evaluate the agreement between clinical diagnosis and autopsy histopathological findings in the cause of death in neonates. Methods: This is a prospective, single-center study of 100 neonatal autopsies conducted at Kasturba Medical College, Manipal which commenced after the approval from the Institutional Ethics Committee and consent from parents. A comprehensive autopsy examination with histopathology examination was performed to determine the cause of death. Ancillary tests and medical records were reviewed by experts for diagnostic confirmation and were coded using the WHO application of ICD-10 to deaths during the perinatal period (ICD-PM). The Goldman criteria were used to compare clinical diagnosis with autopsy findings. Results: Discrepancies between clinical and pathological diagnoses were observed in 61% of cases. Among these, 30% were categorized as class I and 17% as class II, based on the Goldman criteria. An absolute agreement between clinical and pathological diagnoses was noted in 37% of cases. Additionally, minor missed diagnoses were identified in 14% of cases, while 2% were classified as uncertain. The study identified major findings such as intraventricular hemorrhage (34%), infections (32%), pulmonary lesions (23%), and congenital malformations (8%). Blood cultures were positive in 51% of the cases and additional infections such as fungal and viral were identified on histopathological examination. Conclusions: This study highlights the critical role of neonatal autopsies in identifying major and minor histopathological findings that contribute to neonatal mortality. The significant discrepancies between clinical and pathological diagnoses underscore the need for comprehensive postmortem evaluations to enhance diagnostic accuracy. By integrating histopathological and microbiological analyses, this study emphasizes the value of autopsies in improving neonatal care and guiding clinical practice.